Hydrogen bonding or stacking interactions in differentiating duplex stability in oligonucleotides containing synthetic nucleoside probes for alkylated DNA

Article abstract of DOI:10.1002/chem.201204593

Oligonucleotides that hybridize to modified DNA are useful chemical tools to probe the noncovalent interactions that stabilize DNA duplexes. In an effort to better understand the interactions that influence the specificity of hybridization probes for O

Full text of DOI:10.1002/chem.201204593

DOI: 10.1002/chem.201204593
Hydrogen Bonding or Stacking Interactions in Differentiating Duplex
Stability in Oligonucleotides Containing Synthetic Nucleoside Probes for
Alkylated DNA
Hailey L. Gahlon and Shana J. Sturla*[a]
Abstract: Oligonucleotides that hybrid-
ize to modified DNA are useful chemi-
cal tools to probe the noncovalent in-
teractions that stabilize DNA duplexes.
In an effort to better understand the
interactions that influence the specifici-
ty of hybridization probes for O6-alkyl-
guanine lesions, we examined a series
of synthetic nucleoside analogues
(BIM, Benzi, and Peri) with respect to
their ability to stabilize duplex DNA
comprised of native or damaged DNA
oligonucleotides. The base-modified
nucleoside analogues contained system-
atically varied hydrogen-bonding and
p-stacking properties. The nucleoside
probes were incorporated into DNA
and paired opposite canonical bases
(A, T, C, or G), O6-methylguanine (O6-
MeG), O6-benzylguanine (O6-BnG), or
a stable abasic site analogue (tetrahy-
drofuran, THF). On the basis of the
free energy of duplex formation, the
highest degree of stabilization was ob-
served when Peri was paired opposite
O6-MeG. The thermodynamic data sug-
gest that the smaller probes stabilize
DNA duplexes more through hydrogen
bonding, whereas the larger probes,
with a greater capacity to p stack, con-
tribute to duplex stabilization more on
the basis of base stacking. These results
demonstrate that increased helix stabil-
ity could be achieved when BIM,
Benzi, or Peri were paired opposite
damage-containing DNA rather than
unmodified DNA (that is, O6-MeG
rather than G). This knowledge is ex-
pected to be useful in the design and
development of nucleoside analogues
for uses in DNA-based technologies.
Keywords: DNA damage · DNA
recognition · nitrogen heterocycles ·
nucleobases · oligonucleotides
Introduction
detecting low-abundance DNA adducts. Moreover, synthetic
nucleosides can be useful therapeutic and diagnostic tools,
for example, as DNA aptamers that bind and specifically
recognize various biological targets or small molecules.[5]
One example is an abasic site-containing DNA aptamer re-
ported to detect small molecules (for example, adenosine
was detected with a 1–2 mm limit of detection (LOD)).[6]
Evaluation of the factors that modulate DNA stability (that
is, hydrogen bonding, pstacking, solvation, and base-pair
size and shape complementarity) is important for the ration-
al design of nucleoside analogues in DNA-based technolo-
gies.
There are many examples of synthetic base surrogates in
DNA that impart duplex stability, most of which are de-
signed to form stable self-pairs or non-natural heteropairs in
duplex DNA.[7] One example includes synthetic bases that
are size expanded (xDNA) and larger than natural DNA
bases by 2.4 ꢀ. xDNA has recently been shown to encode a
green-fluorescent protein (GFP)-labeled protein that was in-
serted into plasmids containing xDNA in Escherichia coli.[8]
A six-letter genetic alphabet has been created on the basis
of an unnatural base pair that relies on hydrophobic interac-
tions.[9] In contrast to the complementary pairing of synthet-
ic bases in DNA duplexes, however, there are extremely
limited examples of nucleoside analogues that selectively
pair with physiological DNA adducts.[10] For example, the
effect of base-pairing stability in oligonucleotides containing
the DNA adduct 8-oxo-2’-deoxyguanosine has been studied
Noncovalent interactions like hydrogen bonding and base
stacking are important factors that contribute to the thermo-
dynamic stability of DNA duplexes. Chemical modification
of DNA can impact duplex stability by altering the hydro-
gen-bonding or sterics/stacking properties of nucleobases.[1]
Generation of O6-alkylguanine adducts in DNA results in
hydrogen-bonding distortions if the modified base is paired
opposite C.[2] During polymerase-mediated lesion bypass
across O6-alkylguanine lesions, G-to-A transition mutations
can occur.[3] Given the biological significance of O6-alkyl-
guanine adducts and their potential as chemical biomarkers
for carcinogenesis, it is important to better understand the
influence of O6-guanine alkylation on DNA properties and
functions.[4]
Oligonucleotides containing synthetic nucleoside probes,
as specific pairing partners for physiologically relevant DNA
adducts, may be useful tools to elucidate mechanisms of
noncovalent interactions in DNA or to serve as a basis for
[a] H. L. Gahlon, Prof. S. J. Sturla
Department of Health Sciences and Technology
ETH Zꢁrich, Schmelzbergstrasse 9, 8092 Zꢁrich (Switzerland)
Fax : (+41)44-632-1123
E-mail: sturlas@ethz.ch
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201204593.
11062
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 11062 – 11067

FULL PAPER
by using 8-haloguanines as chemical probes.[11] The data ob-
tained supported a model in which the anti conformation of
dG is disrupted, due to the steric bulk at the C8 position.
We have reported previously that a nucleoside derived from
perimidinone (Per; Figure 1A) stabilizes oligonucleotides
when paired opposite the adduct O6-BnG, relative to the
stability of duplexes containing O6-BnG:C (Table 1).[12]
However, to date, there has been no further evaluation of
how structural changes to Per contribute to duplex stability.
In this study, we prepared and evaluated a series of nu-
cleoside probes with varied hydrogen-bonding potential and
base-stacking capabilities (Figure 1A) to test how these fac-
tors may modulate duplex stabilization of the probe:O6-al-
kylguanine pair. Peri is similar in size to Per but lacks the
potential hydrogen-bond-acceptor carbonyl group and hy-
drogen-bond-donor amino proton. BIM and Benzi are
smaller analogues lacking the naphthalene moiety that is
present in Peri and Per. These nucleobase surrogates were
synthesized and incorporated into oligonucleotides. The re-
sulting oligonucleotides were hybridized with complementa-
ry strands containing native or damaged DNA. The DNA
duplexes prepared in this study were characterized on the
basis of their thermodynamic stability. Circular dichroism
measurements were performed to test the effect, on the
global DNA conformation, of incorporating the synthetic
probes opposite damaged DNA in the oligonucleotide du-
plexes. The resulting data reveal how helix stabilization can
be modulated by differences in base-pairing interactions
(that is, hydrogen bonding and p stacking) between oligonu-
cleotides containing nucleoside analogues that pair opposite
O6-alkylguanine lesions.
Figure 1. Modified DNA bases investigated herein: A) Synthetic base sur-
rogates BIM, Benzi, Peri, and Per; B) DNA lesions O6-MeG and O6-
BnG and the abasic site analogue THF.
Results
Influence of probe structure on unmodified duplex stability:
Melting temperatures (Tm values) were compared for du-
plexes containing BIM, Benzi, or Peri paired opposite A, T,
C, or G (Table 2). In comparison to BIM or Benzi, Peri pro-
vided the lowest amount of destabilization (DTm values)
when paired opposite all four native DNA bases, with the
exception of G:Benzi (Table 2). Generally, DNA duplexes
containing BIM, Benzi, or Peri opposite A were less destabi-
lized than those that placed the synthetic bases opposite C,
T, or G. The duplex containing the base pair A:Peri exhibit-
ed a Tm value of 56.78C (8 mm), which was the highest Tm
value observed for duplexes containing a synthetic base
paired with an unmodified base. In contrast, duplexes con-
taining BIM, Benzi, or Peri opposite C tended to have the
Table 1. Tm values for 15-mer duplexes containing 5’-CCGPTATACC-
GACAA-3’ (P: Per) and 5’-TTGTCGGTATANCGG-3’ (N: G, O6-MeG,
or O6-BnG). The unmodified G:C pair has a Tm value of (64.3Æ0.3)8C.
Measurements were made in aqueous buffer solution (250 mm NaCl,
0.2 mm ethylenediaminetetraacetate (EDTA), 20 mm phosphate buffer,
pH 7.0).
Base pair (N:P)
Tm [8C]
G:Per
55.4Æ0.7
57.4Æ0.6
56.8Æ0.6
53.4Æ0.6
58.0Æ0.3
54.6Æ0.2
O6-MeG:Per
O6-BnG:Per
THF:Per
O6-MeG:C
O6-BnG:C
Table 2. Tm values for duplexes containing nucleoside probes BIM, Benzi, and Peri paired opposite native or damaged DNA strands.[a,b]
N:P[c]
Tm
[8C]
DTm
N:P[c]
Tm
[8C]
DTm
N:P[c]
Tm
[8C]
DTm
[8C][d]
[8C][d]
[8C][d]
A:BIM
T:BIM
C:BIM
G:BIM
53.1Æ0.2
50.2Æ0.3
47.1Æ0.2
52.1Æ0.2
56.0Æ0.3
53.6Æ0.2
50.2Æ0.3
À9.4
À12.7
À16.9
À12.2
À2.0
À1.0
–
A:Benzi
T:Benzi
C:Benzi
G:Benzi
55.1Æ0.3
52.9Æ0.3
50.3Æ0.3
54.0Æ0.3
58.6Æ0.4
56.2Æ0.1
51.0Æ0.2
À7.4
À10.0
À13.7
À10.3
+0.6
+1.6
–
A:Peri
T:Peri
C:Peri
G:Peri
56.7Æ0.6
53.7Æ0.5
53.4Æ0.6
53.4Æ0.6
57.7Æ0.6
57.7Æ0.6
53.4Æ0.1
À5.8
À9.2
À10.6
À10.9
À0.3
+3.1
–
O6-MeG:BIM
O6-BnG:BIM
THF:BIM
O6-MeG:Benzi
O6-BnG:Benzi
THF:Benzi
O6-MeG:Peri
O6-BnG:Peri
THF:Peri
[a] Data reported for 8 mm duplexes in aqueous buffer solution (250 mm NaCl, 0.2 mm EDTA, 20 mm phosphate buffer, pH 7.0). [b] The unmodified
duplex (G:C pair) has Tm =(64.3Æ0.3)8C. [c] DNA sequences comprise 5’-CCGPTATACCGACAA-3’ (P: BIM, Benzi, or Peri) and 5’-TTGTCGGTA-
TANCGG-3’ (N: G, O6-MeG, or O6-BnG). [d] DTm [8C] values are defined as the difference between the Tm values of duplexes with canonical bases
versus the corresponding N:P pairs: A:P to A:T (62.58C), T:P to T:A (62.98C), C:P to C:G (64.08C), G:P to G:C (64.38C), O6-MeG:P to O6-MeG:C
(58.08C), and O6-BnG:P to O6-BnG:C (54.68C).
Chem. Eur. J. 2013, 19, 11062 – 11067
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
11063

S. J. Sturla and H. L. Gahlon
Table 5. Thermodynamic parameters for Benzi-containing duplexes.
lowest Tm values; the base pair of C:BIM was the least
stable of the series (Tm =47.18C). Furthermore, in compari-
son to the corresponding C:G duplex, C:BIM was greatly
destabilized (DTm =À16.98C).
N:P
DH8 [kcalmolÀ1
]
DS8 [calKÀ1 molÀ1
]
DG8 [kcalmolÀ1
]
A:Benzi
T:Benzi
C:Benzi
G:Benzi
À87.3Æ13.8
À83.5Æ12.7
À83.7Æ6.5
À86.4Æ5.9
À246Æ42.6
À233Æ38.9
À235Æ20.1
À239Æ18.2
À256Æ24.0
À246Æ12.2
À230Æ63.7
À14.0Æ0.04
À14.1Æ0.05
À13.7Æ0.02
À15.2Æ0.02
À16.9Æ0.02
À15.7Æ0.01
À14.0Æ0.08
Concentration-dependent changes in the DNA-duplex Tm
value reflect duplex stability (DG8), as well as entropic
(DS8) and enthalpic (DH8) parameters.[13] Thermodynamical-
ly, A:BIM and A:Benzi duplexes have similar enthalpy and
entropy values (À86.3 kcalmolÀ1, À241 calKÀ1 molÀ1 and
À87.3 kcalmolÀ1, À246 calKÀ1 molÀ1, respectively), whereas
an A:Peri duplex has a less favorable DH8 value (À68.9 kcal
O6-MeG:Benzi À93.2Æ7.9
O6-BnG:Benzi À89.0Æ4.0
THF:Benzi
À82.5Æ22.0
molÀ1) but
a
slightly more favorable DS8 value
alkylG:probe series, O6-MeG:Benzi, O6-BnG:Benzi, and O6-
BnG:Peri were less destabilized than the corresponding O6-
alkylG:C duplexes (DTm = +0.6, +1.6, and +3.18C, respec-
tively). Duplexes containing BIM opposite O6-MeG or O6-
BnG were more destabilized than the O6-MeG:C and O6-
BnG:C duplexes (DTm =À2.0 and À1.08C, respectively;
Table 2). The highest DNAduplex stability observed for the
synthetic probes paired opposite O6-BnG was for O6-
BnG:Peri (Tm =57.78C, Table 2; DG8=À17.4 kcalmolÀ1,
Table 3). The presence of modified bases in DNA may alter
the global DNA conformation. Therefore, circular dichroism
experiments were performed with duplexes containing BIM,
Benzi, or Peri opposite G, O6-MeG, O6-BnG, or an abasic
site analogue (THF). All of the duplexes appeared to form
a right-handed double-helical structure (B DNA) and the
CD data suggested that the global structure of the DNA was
not perturbed (see Figure S5–S7 in the Supporting Informa-
tion).
(À183 calKÀ1 molÀ1). In general, it appeared that duplex sta-
bilization diminished if the smaller aromatic probes BIM
and Benzi were paired opposite the pyrimidine bases T and
C. The T:Benzi and C:Benzi duplexes exhibited Tm values
of 52.9 and 50.38C, respectively. Both were more stable than
the T:BIM and C:BIM duplexes by 2.7 and 3.28C, respec-
tively. This trend did not hold for Peri, because similar Tm
values resulted when Peri was paired opposite T, C, or G
(53.7, 53.4, and 53.48C, respectively).
Influence of probe structure on duplex stability and helix
conformation when paired with O6-alkylguanine: Duplexes
comprised of damaged strands containing O6-MeG or O6-
BnG and complementary strands containing the synthetic
bases BIM, Benzi, and Peri were evaluated by thermal-de-
naturation analysis to assess relative stabilities (Table 3,
Table 3. Thermodynamic parameters for Peri-containing duplexes.
Influence of probe structure on stability of DNA duplexes
with an abasic site: We evaluated the stability of DNA du-
plexes containing an abasic site on the basis of the chemical-
ly stable abasic site analogue THF (Figure 1B). The Tm
values for the THF:BIM (50.28C) and THF:Benzi (51.08C)
duplexes suffered from the highest degree of duplex destabi-
lization in the probe series paired opposite THF (Table 2).
The enthalpic and entropic values were similar for
N:P
DH8 [kcalmolÀ1
]
DS8 [calKÀ1 molÀ1
]
DG8 [kcalmolÀ1
]
A:Peri
T:Peri
C:Peri
G:Peri
À68.9Æ5.5
À94.1Æ22.2
À77.3Æ6.0
À96.0Æ14.4
À183Æ16.9
À261Æ67.5
À211Æ19.1
À268Æ43.6
À308Æ53.5
À272Æ15.6
À270Æ37.1
À14.5Æ0.03
À16.4Æ0.06
À14.5Æ0.30
À16.2Æ0.30
À18.4Æ0.03
À17.4Æ0.02
À15.9Æ0.03
O6-MeG:Peri À110Æ18.1
O6-BnG:Peri À98.3Æ4.9
THF:Peri
À96.4Æ9.3
THF:BIM
(DH8=À82.1 kcalmolÀ1,
DS8=
À229 calKÀ1 molÀ1) and THF:Benzi (DH8=À82.5 kcalmolÀ1,
DS8=À230 calKÀ1 molÀ1) duplexes (Table 4 and Table 5, re-
spectively). A more stable Tm value of 53.48C was observed
for THF:Peri than those with BIM and Benzi. Additionally,
the THF:Peri duplex displayed a larger negative DG8 value
(À15.9 kcalmolÀ1) relative to those with BIM or Benzi,
which suggests that the larger p surface area of Peri contrib-
utes to increased stability opposite THF. For the same oligo-
nucleotide sequences, the Tm values for DNA containing
THF opposite A, G, T, and C were 50.7, 46.4, 47.3, and
48.48C, respectively.
Table 4. Thermodynamic parameters for BIM-containing duplexes.
N:P
DH8 [kcalmolÀ1
]
DS8 [calKÀ1 molÀ1
]
DG8 [kcalmolÀ1
]
A:BIM
T:BIM
C:BIM
G:BIM
À86.3Æ10.8
À83.2Æ7.2
À84.1Æ8.4
À85.5Æ7.4
À241Æ33.2
À232Æ22.7
À242Æ26.6
À238Æ21.2
À256Æ25.5
À248Æ28.4
À229Æ52.8
À14.1Æ0.03
À14.5Æ0.03
À12.0Æ0.03
À14.6Æ0.02
À16.2Æ0.02
À15.2Æ0.03
À13.9Æ0.06
O6-MeG:BIM À92.5Æ8.8
O6-BnG:BIM À89.1Æ9.3
THF:BIM
À82.1Æ17.1
Table 4, and Table 5). The most stable pair within this series,
on the basis of the Tm values, was O6-MeG:Benzi (Tm =
58.68C). However, based on the free energy of duplex for-
mation, O6-MeG:Peri had the largest negative DG8 value of
À18.4 kcalmolÀ1 (versus À16.9 kcalmolÀ1 for O6-MeG:Ben-
zi; Table 3 and Table 5, respectively). Moreover, in the O6-
Discussion
Hydrogen bonding and base stacking contribute to the sta-
bility of duplex DNA.[14] Hydrogen bonding is a significant
driving force for DNA-mediated molecular assembly and
11064
www.chemeurj.org
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 11062 – 11067

Duplex Stability in Oligonucleotides Containing Synthetic Nucleoside Probes
FULL PAPER
contributes around 1.8 kcalmolÀ1 per hydrogen bond.[15]
Base stacking imparts DNA stability from overlapping p
systems, and neighboring base pairs can make contacts up to
115 ꢀ2.[16] The nucleoside probes in this study were designed
to differ in their hydrogen-bonding potential, as well as
base-stacking propensities. BIM and Benzi (Figure 1) have
different hydrogen-bonding capacities due to the presence
of the carbonyl moiety in Benzi; however, they have similar
p-stacking capacities. Peri has similar hydrogen-bonding po-
tential to that of BIM, but Peri is larger and has a greater
propensity to p stack. Per has a similar size and stacking
propensity to Peri, but its hydrogen-bonding potential is like
that of Benzi. Thus, the present study provides a systematic
chemical basis for understanding noncovalent interactions
that contribute to duplex stability in DNA duplexes between
strands containing O6-alkylguanine and complementary se-
quences containing nucleoside probes.
T, C, or G. In contrast, duplexes containing BIM or Benzi
paired opposite T, C, or G displayed significant differences
in Tm values. For example, a comparison of the Tm values of
duplexes with C:BIM (47.18C) versus G:BIM (52.18C) re-
vealed a DTm value of 5.08C. This DTm value may be ex-
plained 1) by the potential variation in hydrogen bonding
between C:BIM and G:BIM and 2) by the increased intra-
strand stacking potential of G. Moreover, the Tm values for
C:BIM and G:Benzi duplexes differ by 6.98C. With the as-
sumption that BIM and Benzi have similar p-stacking abili-
ties, these significant Tm differences suggest that hydrogen-
bonding differences between the two smaller probes (BIM
versus Benzi) may influence DNA-duplex stability to a
larger extent. Therefore, we hypothesize that hydrogen
bonding is a more relevant factor in increasing DNA-duplex
stability for the smaller probes (BIM and Benzi), whereas
base stacking seems to be the main contributor to duplex
stabilization for the larger naphthalene probes (Peri and
Per).
Duplexes containing nucleoside probes paired opposite
canonical bases were the least destabilized, on the basis of
the DTm values, when the probes were paired opposite A
(Figure 2). This trend could be due to the fact that, amongst
An interesting finding is that duplexes containing O6-
MeG:Benzi or O6-MeG:Per have Tm values that are similar
(58.68C and 57.48C). These Tm values are comparable to
those for duplexes containing O6-MeG paired opposite C
(58.08C). On the basis of a reported crystal structure of an
O6-MeG:C pair within a polymerase active site, this combi-
nation forms a wobble base pair.[18] By analogy, O6-MeG:-
Benzi and O6-MeG:Per may orient similarly (Figure 3). This
Figure 2. Melting temperatures for the BIM, Benzi, and Peri probes op-
posite native DNA bases. DNA sequences comprise 15-mer duplexes
containing 5’-CCGPTATACCGACAA-3’ (P: BIM, Benzi, or Peri) and
5’-TTGTCGGTATANCGG-3’ (N: native DNA bases).
the four canonical bases, A has the greatest base-stacking
ability in DNA, for which A>G>T=C.[17] Furthermore,
A:Peri was the least destabilized (DTm =À5.88C) and had
the most favorable DG8 value (À14.5 kcalmolÀ1) for the
A:probe oligonucleotide series. The combination of the
stacking potential of A and Periꢃs strong potential to base
stack may explain the increased stabilization relative to du-
plexes containing the smaller BIM and Benzi probes when
paired opposite A.
Duplexes comprised of oligonucleotides placing Peri op-
posite T, C, or G had similar Tm values (Figure 2), which
suggests that base stacking, rather than hydrogen bonding,
may be the main factor driving DNA-duplex stability for
Peri. It appears that, if hydrogen bonding were a more sig-
nificant contributing factor, a greater difference in the Tm
values may have been expected when it was placed opposite
Figure 3. Schematic rendering of the potential hydrogen bonds in A) the
O6-MeG:C wobble base pair; B) O6-MeG:Benzi; and C) O6-MeG:Per.
model would suggest that deleting the NH hydrogen-bond
donor on the Benzi and Per probes would depress the melt-
ing temperatures of the corresponding duplexes (the NH
donors on Benzi and Per are circled in Figure 3).
With respect to thermodynamic parameters derived for
the DNA duplexes in this study, enthalpy–entropy compen-
sation can present moderate changes in the overall free
energy.[19] Nevertheless, a general trend emerged, in which
Chem. Eur. J. 2013, 19, 11062 – 11067
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
11065

S. J. Sturla and H. L. Gahlon
parameters for duplex melting were determined by the Vanꢃt Hoff
method from linear plots of 1/Tm versus ln[CT] by measuring the Tm
value as a function of concentration; the final duplex concentrations
were 8.0, 4.0, 2.8, 1.4, and 0.8 mm (see the Supporting Information for the
Vanꢃt Hoff plots).[13]
the smaller probes (BIM and Benzi) seemed to contribute
more to DNA-duplex stability from enthalpic contributions
(for example, hydrogen bonding). In contrast, the larger
naphthalene probes (Peri and Per) seemed to contribute
more to DNA-duplex stability on the basis of entropic ef-
fects (for example, p stacking).
Circular dichroism measurements: CD spectra were recorded on
a
JASCO J-815 spectropolarimeter equipped with a Peltier temperature
controller. Oligodeoxynucleotides containing G, O6-BnG, O6-MeG, or the
abasic site analogue THF were annealed with a complementary strand
containing Benzi, BIM, or Peri in a buffer solution (250 mm NaCl,
0.2 mm EDTA, 20 mm phosphate buffer, pH 7.0) by heating to 908C and
cooling to 258C overnight. For analysis, 2 mL aliquots of 2.8 mm duplex
solutions were used. The samples were scanned from 320–200 nm at
258C. A wavelength step of 0.1 nm in a 10 mm pathlength quartz cuvette
and a digital integration time of 1 s were used. Spectra were obtained as
an average of 3–5 scans.
Conclusion
The results of this study demonstrate that a series of DNA
adduct-targeting nucleoside probes stabilize native and
lesion-containing DNA to different extents depending on
their hydrogen-bonding or p-stacking abilities. On the basis
of the free energy of duplex formation, the greatest stabili-
zation involving oligonucleotides with the promutagenic
lesion O6-alkylguanine was observed when Peri was placed
opposite O6-MeG. The data suggest that smaller probes
(BIM and Benzi) may influence DNA stability largely on
the basis of hydrogen-bonding interactions, whereas p-stack-
ing may be more significant for duplex stabilization for the
naphthyl-based Peri and Per probes. Optimization of these
synthetic nucleosides is envisioned to contribute to the
chemical diversity of synthetic base surrogates used in
DNA-based technologies and in the development of syn-
thetic oligonucleotides as hybridization probes.
Acknowledgements
This work was supported by funding from the European Research Coun-
cil (grant no. 260341). We thank Alexandru Zabara and Prof. Raffaele
Mezzenga (ETH Zꢁrich) for help in acquiring the circular dichroism
spectra. Thanks to Rahul Lad for contributions to the synthesis and ther-
mal stability measurements in this study.
[1] a) T. Zaliznyak, R. Bonala, F. Johnson, C. de Los Santos, Chem. Res.
Toxicol. 2006, 19, 745–752; b) H. Ling, J. M. Sayer, B. S. Plosky, H.
Yagi, F. Boudscoq, R. Woodgate, D. M. Jerina, W. Yang, Proc. Natl.
Acad. Sci. USA 2004, 101, 2265–2269; c) M. Lukin, T. Zaliznyak, F.
Johnson, C. de Los Santos, Nucleic Acids Res. 2012, 40, 2759–2770.
[2] R. L. Eoff, K. C. Angel, M. Egli, F. P. Guengerich, J. Biol. Chem.
2007, 282, 13573–13584.
Experimental Section
[3] G. T. Pauly, R. C. Moschel, Chem. Res. Toxicol. 2001, 14, 894–900.
[4] a) F. P. Guengerich, Chem. Rev. 2006, 106, 420–452; b) K. S. Gates,
Chem. Res. Toxicol. 2009, 22, 1747–1760; c) T. Shibuya, K. Morimo-
to, Mutat. Res. 1993, 297, 3–38; d) G. Boysen, B. F. Pachkowski, J.
Nakamura, J. A. Swenberg, Mutat. Res. 2009, 678, 76–94; e) J. S. H.
Yoo, F. P. Guengerich, C. S. Yang, Cancer Res. 1988, 48, 1499–1504;
f) J. L. Tubbs, A. E. Pegg, J. A. Tainer, DNA Repair 2007, 6, 1100–
1115; g) B. Singer, Cancer Invest. 1984, 2, 233–238.
[5] a) F. Tolle, G. Mayer, Chem. Sci. 2013, 4, 60–67; b) E. W. Orava, N.
Cicmil, J. Gariepy, Biochim. Biophys. Acta Biomembr. 2010, 1798,
2190–2200; c) C. L. A. Hamula, H. Zhang, F. Li, Z. Wang, X. C. Le,
X. F. Li, Trends Anal. Chem. 2011, 30, 1587–1597.
[6] S. Nishizawa, Y. Sato, Z. Xu, K. Morita, M. Li, N. Teramae, Supra-
mol. Chem. 2010, 22, 467–476.
[7] a) A. M. Leconte, G. T. Hwang, S. Matsuda, P. Capek, Y. Hari, F. E.
Romesberg, J. Am. Chem. Soc. 2008, 130, 2336–2343; b) I. Hirao,
M. Kimoto, Proc. Jpn. Acad. Ser. B 2012, 88, 345–367.
[8] A. T. Krueger, L. W. Peterson, J. Chelliserry, D. J. Kleinbaum, E. T.
Kool, J. Am. Chem. Soc. 2011, 133, 18447–18451.
[9] Y. Wu, A. K. Ogawa, M. Berger, D. L. McMinn, P. G. Schultz, F. E.
Romesberg, J. Am. Chem. Soc. 2000, 122, 7621–7632.
[10] a) O. Nakagawa, S. Ono, Z. Li, A. Tsujimoto, S. Sasaki, Angew.
Chem. 2007, 119, 4584–4587; Angew. Chem. Int. Ed. 2007, 46, 4500–
4503; b) T. J. Matray, E. T. Kool, J. Am. Chem. Soc. 1998, 120, 6191–
6192.
[11] M. L. Hamm, S. Rajguru, A. M. Downs, R. Cholera, J. Am. Chem.
Soc. 2005, 127, 12220–12221.
Synthetic bases: Phosphoramidite deriviatives of Per, Peri, BIM, and
Benzi were synthesized and incorporated into 15-mer DNA oligonucleo-
tides by solid-phase DNA synthesis (for detailed procedures, see the Sup-
porting Information).
Oligodeoxynucleotides: Oligonucleotides were synthesized on a Mer-
made 4 DNA synthesizer (Bioautomation Corporation) on the basis of
standard b-cyanoethylphosphoramidite chemistry. Stepwise coupling
yields for oligonucleotides containing BIM, Benzi, and Peri were greater
than 85%, as determined by trityl cation absorbance. Oligodeoxynucleo-
tides were synthesized in trityl-off mode on the 5’-end. Cleavage from
the controlled pore glass (CPG) solid support and final deprotection
were achieved by treatment with 30% aqueous ammonium hydroxide at
558C for 16 h. The resulting crude oligodeoxynucleotides were purified
by reversed-phase HPLC on a Luna 5 mm C18 (2) 100 ꢀ 250ꢄ10 mm
column (Phenomenex). The chromatography mobile phases were solvent
A (15 mm aqueous ammonium acetate (pH 7.0) or 50 mm triethylammo-
nium acetate (pH 7.0)) and solvent B (acetonitrile). The solvent gradient
started with 95% A, increased linearly to 95% B over 20 min, and was
held at 95% B for 20 min. Fractions corresponding to the oligodeoxynu-
cleotide peak were collected and lyophilized, and the resulting pure (>
95%) oligodeoxynucleotides were stored at À208C.
Thermal denaturation studies: Stock solutions of oligodeoxynucleotides
were prepared in aqueous buffer solution (250 mm NaCl, 0.2 mm EDTA,
20 mm phosphate buffer, pH 7.0). Melting-temperature analyses were car-
ried out in Teflon-stoppered 1 cm pathlength quartz cells on a Varian
Cary 100 Bio UV/Vis spectrophotometer equipped with a thermal pro-
grammer. Duplex solutions (200–400 mL) were heated to 808C and al-
lowed to slowly cool to 208C at a rate of 18CminÀ1. UV absorbance was
monitored at 260 nm. Melting temperatures (Tm) were determined by the
first-derivative method with the recalculation application in the Cary
Thermal software. All measurements were performed in triplicate, and
the data is presented as the meanÆstandard deviation. Thermodynamic
[12] J. Gong, S. J. Sturla, J. Am. Chem. Soc. 2007, 129, 4882–4883.
[13] L. A. Marky, K. J. Breslauer, Biopolymers 1987, 26, 1601–1620.
[14] a) I. R. Gould, P. A. Kollman, J. Am. Chem. Soc. 1994, 116, 2493–
2499; b) E. A. Meyer, R. K. Castellano, F. Diederich, Angew. Chem.
2003, 115, 1244–1287; Angew. Chem. Int. Ed. 2003, 42, 1210–1250.
[15] H. Klump, T. Ackermann, Biopolymers 1971, 10, 513–522.
11066
www.chemeurj.org
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 11062 – 11067

Duplex Stability in Oligonucleotides Containing Synthetic Nucleoside Probes
FULL PAPER
[16] E. T. Kool, J. C. Morales, K. M. Guckian, Angew. Chem. 2000, 112,
1046–1068; Angew. Chem. Int. Ed. 2000, 39, 990–1009.
[17] K. M. Guckian, B. A. Schweitzer, R. X. F. Ren, C. J. Sheils, D. C.
Tahmassebi, E. T. Kool, J. Am. Chem. Soc. 2000, 122, 2213–2222.
[18] R. L. Eoff, A. Irimia, M. Egli, F. P. Guengerich, J. Biol. Chem. 2006,
282, 1456–1467.
[19] a) E. B. Starikov, B. Norden, J. Phys. Chem. B 2009, 113, 4698–4707;
b) E. B. Starikov, B. Norden, J. Phys. Chem. B 2009, 113, 11375–
11377.
Received: December 26, 2012
Revised: April 17, 2013
Published online: June 25, 2013
Chem. Eur. J. 2013, 19, 11062 – 11067
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
11067